Coronary Artery Surgery in New York State

To the Editor.— We read with great interest the article by Hannan et al. 1 We commend the authors for the attempt to establish a risk stratification method by which hospitals can compare results. Several aspects of the article deserve comments and raise questions. The basic premise of the method is to define risk-stratified mortality rates. Yet in Table 5 of the Hannan et al article (p 2772), a risk-stratified column was left out, though it was included in a preliminary draft sent to all hospitals in New York State. The risk-adjusted mortality from that preliminary draft for Table 5 is depicted in the Table. The important emphasis of the article lies in the comparison of quality of care given by hospitals, using standardization of mortality rates. Ranking of hospitals by crude mortality rates is not only uninformative, but it is misleading in that it does not take into account